Semantics Driven Agent Programming

In the last two decades the Information Technology changed substantially the life of people all around the World. Just a few years ago, for instance, paper support was needed to exchange all kind of data, while now electronics is indeed the main instrument of communication. This mutation was originally due mostly to the efficiency, while now it is, hopefully, also due to an increased attention towards environmental issues. Information and data have proved over the time the importance of their role, contributing to a plethora of applications that allow the physical world to interact with mankind by the means of services dispatched pervasively and freely accessible. The Internet is the kernel of such complex setup that is called Internet of Things (IoT). The IoT inherited from the Internet a chaotic interface. Protocols, conventions, mechanisms are different from an application to the other, and it is difficult and expensive to discover and make applications compatible with one another. From this consideration two exceptional ideas were born, namely the Semantic Web and the Web of Things (WoT). The latter would unify the IoT on an application level shared view, enabling standard discovery mechanisms and definitions. The former, on the other hand, intents to provide the tools to formalize the knowledge contents of the World Wide Web in a simultaneously human and machine understandable way. This Thesis aims to explore both these two concepts and merge them into the Semantic Web of Things using the best of each. Therefore we hereby propose, describe, evaluate and use two ontologies: the Internet of Musical Things ontology, aiming to outline a semantic description of IoT; and a Semantic WoT ontology, aiming to push further the state of the art of IoT unification and standardization through a dynamic semantic approach.